summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorDavid Green <david.green@arm.com>2018-07-28 08:20:10 +0000
committerDavid Green <david.green@arm.com>2018-07-28 08:20:10 +0000
commitfc4b0fe0a2c3a0908afdb625f0cef843acdc126c (patch)
tree834b0cb3b5376bed6c955ada249a426dbf3e38d2
parentf947608ddfe026eaf013524e5fe09d3c425a5cf0 (diff)
downloadbcm5719-llvm-fc4b0fe0a2c3a0908afdb625f0cef843acdc126c.tar.gz
bcm5719-llvm-fc4b0fe0a2c3a0908afdb625f0cef843acdc126c.zip
[GlobalOpt] Test array indices inside structs for out-of-bounds accesses
We now, from clang, can turn arrays of static short g_data[] = {16, 16, 16, 16, 16, 16, 16, 16, 0, 0, 0, 0, 0, 0, 0, 0}; into structs of the form @g_data = internal global <{ [8 x i16], [8 x i16] }> ... GlobalOpt will incorrectly SROA it, not realising that the access to the first element may overflow into the second. This fixes it by checking geps more thoroughly. I believe this makes the globalsra-partial.ll test case invalid as the %i value could be out of bounds. I've re-purposed it as a negative test for this case. Differential Revision: https://reviews.llvm.org/D49816 llvm-svn: 338192
-rw-r--r--llvm/lib/Transforms/IPO/GlobalOpt.cpp118
-rw-r--r--llvm/test/Transforms/GlobalOpt/globalsra-multigep.ll16
-rw-r--r--llvm/test/Transforms/GlobalOpt/globalsra-partial.ll5
3 files changed, 66 insertions, 73 deletions
diff --git a/llvm/lib/Transforms/IPO/GlobalOpt.cpp b/llvm/lib/Transforms/IPO/GlobalOpt.cpp
index 1af7e689477..1761d7faff5 100644
--- a/llvm/lib/Transforms/IPO/GlobalOpt.cpp
+++ b/llvm/lib/Transforms/IPO/GlobalOpt.cpp
@@ -357,6 +357,41 @@ static bool CleanupConstantGlobalUsers(Value *V, Constant *Init,
return Changed;
}
+static bool isSafeSROAElementUse(Value *V);
+
+/// Return true if the specified GEP is a safe user of a derived
+/// expression from a global that we want to SROA.
+static bool isSafeSROAGEP(User *U) {
+ // Check to see if this ConstantExpr GEP is SRA'able. In particular, we
+ // don't like < 3 operand CE's, and we don't like non-constant integer
+ // indices. This enforces that all uses are 'gep GV, 0, C, ...' for some
+ // value of C.
+ if (U->getNumOperands() < 3 || !isa<Constant>(U->getOperand(1)) ||
+ !cast<Constant>(U->getOperand(1))->isNullValue())
+ return false;
+
+ gep_type_iterator GEPI = gep_type_begin(U), E = gep_type_end(U);
+ ++GEPI; // Skip over the pointer index.
+
+ // For all other level we require that the indices are constant and inrange.
+ // In particular, consider: A[0][i]. We cannot know that the user isn't doing
+ // invalid things like allowing i to index an out-of-range subscript that
+ // accesses A[1]. This can also happen between different members of a struct
+ // in llvm IR.
+ for (; GEPI != E; ++GEPI) {
+ if (GEPI.isStruct())
+ continue;
+
+ ConstantInt *IdxVal = dyn_cast<ConstantInt>(GEPI.getOperand());
+ if (!IdxVal || (GEPI.isBoundedSequential() &&
+ IdxVal->getZExtValue() >= GEPI.getSequentialNumElements()))
+ return false;
+ }
+
+ return llvm::all_of(U->users(),
+ [](User *UU) { return isSafeSROAElementUse(UU); });
+}
+
/// Return true if the specified instruction is a safe user of a derived
/// expression from a global that we want to SROA.
static bool isSafeSROAElementUse(Value *V) {
@@ -374,84 +409,25 @@ static bool isSafeSROAElementUse(Value *V) {
if (StoreInst *SI = dyn_cast<StoreInst>(I))
return SI->getOperand(0) != V;
- // Otherwise, it must be a GEP.
- GetElementPtrInst *GEPI = dyn_cast<GetElementPtrInst>(I);
- if (!GEPI) return false;
-
- if (GEPI->getNumOperands() < 3 || !isa<Constant>(GEPI->getOperand(1)) ||
- !cast<Constant>(GEPI->getOperand(1))->isNullValue())
- return false;
-
- for (User *U : GEPI->users())
- if (!isSafeSROAElementUse(U))
- return false;
- return true;
-}
-
-/// U is a direct user of the specified global value. Look at it and its uses
-/// and decide whether it is safe to SROA this global.
-static bool IsUserOfGlobalSafeForSRA(User *U, GlobalValue *GV) {
- // The user of the global must be a GEP Inst or a ConstantExpr GEP.
- if (!isa<GetElementPtrInst>(U) &&
- (!isa<ConstantExpr>(U) ||
- cast<ConstantExpr>(U)->getOpcode() != Instruction::GetElementPtr))
- return false;
-
- // Check to see if this ConstantExpr GEP is SRA'able. In particular, we
- // don't like < 3 operand CE's, and we don't like non-constant integer
- // indices. This enforces that all uses are 'gep GV, 0, C, ...' for some
- // value of C.
- if (U->getNumOperands() < 3 || !isa<Constant>(U->getOperand(1)) ||
- !cast<Constant>(U->getOperand(1))->isNullValue() ||
- !isa<ConstantInt>(U->getOperand(2)))
- return false;
-
- gep_type_iterator GEPI = gep_type_begin(U), E = gep_type_end(U);
- ++GEPI; // Skip over the pointer index.
-
- // If this is a use of an array allocation, do a bit more checking for sanity.
- if (GEPI.isSequential()) {
- ConstantInt *Idx = cast<ConstantInt>(U->getOperand(2));
-
- // Check to make sure that index falls within the array. If not,
- // something funny is going on, so we won't do the optimization.
- //
- if (GEPI.isBoundedSequential() &&
- Idx->getZExtValue() >= GEPI.getSequentialNumElements())
- return false;
-
- // We cannot scalar repl this level of the array unless any array
- // sub-indices are in-range constants. In particular, consider:
- // A[0][i]. We cannot know that the user isn't doing invalid things like
- // allowing i to index an out-of-range subscript that accesses A[1].
- //
- // Scalar replacing *just* the outer index of the array is probably not
- // going to be a win anyway, so just give up.
- for (++GEPI; // Skip array index.
- GEPI != E;
- ++GEPI) {
- if (GEPI.isStruct())
- continue;
-
- ConstantInt *IdxVal = dyn_cast<ConstantInt>(GEPI.getOperand());
- if (!IdxVal ||
- (GEPI.isBoundedSequential() &&
- IdxVal->getZExtValue() >= GEPI.getSequentialNumElements()))
- return false;
- }
- }
-
- return llvm::all_of(U->users(),
- [](User *UU) { return isSafeSROAElementUse(UU); });
+ // Otherwise, it must be a GEP. Check it and its users are safe to SRA.
+ return isa<GetElementPtrInst>(I) && isSafeSROAGEP(I);
}
/// Look at all uses of the global and decide whether it is safe for us to
/// perform this transformation.
static bool GlobalUsersSafeToSRA(GlobalValue *GV) {
- for (User *U : GV->users())
- if (!IsUserOfGlobalSafeForSRA(U, GV))
+ for (User *U : GV->users()) {
+ // The user of the global must be a GEP Inst or a ConstantExpr GEP.
+ if (!isa<GetElementPtrInst>(U) &&
+ (!isa<ConstantExpr>(U) ||
+ cast<ConstantExpr>(U)->getOpcode() != Instruction::GetElementPtr))
return false;
+ // Check the gep and it's users are safe to SRA
+ if (!isSafeSROAGEP(U))
+ return false;
+ }
+
return true;
}
diff --git a/llvm/test/Transforms/GlobalOpt/globalsra-multigep.ll b/llvm/test/Transforms/GlobalOpt/globalsra-multigep.ll
new file mode 100644
index 00000000000..87a8486d881
--- /dev/null
+++ b/llvm/test/Transforms/GlobalOpt/globalsra-multigep.ll
@@ -0,0 +1,16 @@
+; RUN: opt < %s -globalopt -S | FileCheck %s
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+@g_data = internal unnamed_addr global <{ [8 x i16], [8 x i16] }> <{ [8 x i16] [i16 16, i16 16, i16 16, i16 16, i16 16, i16 16, i16 16, i16 16], [8 x i16] zeroinitializer }>, align 16
+; We cannot SRA here due to the second gep meaning the access to g_data may be to either element
+; CHECK: @g_data = internal unnamed_addr constant <{ [8 x i16], [8 x i16] }>
+
+define i16 @test(i64 %a1) {
+entry:
+ %g1 = getelementptr inbounds <{ [8 x i16], [8 x i16] }>, <{ [8 x i16], [8 x i16] }>* @g_data, i64 0, i32 0
+ %arrayidx.i = getelementptr inbounds [8 x i16], [8 x i16]* %g1, i64 0, i64 %a1
+ %r = load i16, i16* %arrayidx.i, align 2
+ ret i16 %r
+}
diff --git a/llvm/test/Transforms/GlobalOpt/globalsra-partial.ll b/llvm/test/Transforms/GlobalOpt/globalsra-partial.ll
index 6f24128c42b..141ee1bb5a8 100644
--- a/llvm/test/Transforms/GlobalOpt/globalsra-partial.ll
+++ b/llvm/test/Transforms/GlobalOpt/globalsra-partial.ll
@@ -1,11 +1,12 @@
-; In this case, the global can only be broken up by one level.
+; In this case, the global cannot be merged as i may be out of range
; RUN: opt < %s -globalopt -S | FileCheck %s
target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-v64:64:64-v128:128:128"
@G = internal global { i32, [4 x float] } zeroinitializer ; <{ i32, [4 x float] }*> [#uses=3]
-; CHECK-NOT: 12345
+; CHECK: @G = internal unnamed_addr global { i32, [4 x float] }
+; CHECK: 12345
define void @onlystore() {
store i32 12345, i32* getelementptr ({ i32, [4 x float] }, { i32, [4 x float] }* @G, i32 0, i32 0)
ret void
OpenPOWER on IntegriCloud